[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: W3C response to proposed Atom Publishing Format and Protocol (atompub) working group


I very much agree that going to a standards organisation is not automatically a good thing; there are many examples of people trying to use standards orgs to do R&D, and thereby wasting a lot of time and effort.

That said, I think there are good reasons for Atom doing this.

First of all, the Atom Community doesn't have any formal, well-understood way to make decisions. As a result, people feeling disenfranchised when a decision doesn't go their way; and it increases the likelihood that the result we come up with doesn't reflect a shared vision that the entire community has bought into.

While we could get together and come up with a process, there isn't a good reason to do this; established standards organisations like the W3C and the IETF provide a forum with well-understood and extremely well-thought-out rules that have met the test of time. Starting your own consortium or standards org is extremely expensive and time consuming, and fraught with legal complications (to give you just a taste, consider the words "anti-trust").

Secondly, there is currently no formal policies or mechanisms for working with Intellectual Property Rights and Copyright in the Atom community. While these issues haven't bit us yet, there's no guarantee that they won't in the future, especially as Atom gains momentum. By working with people who understand these issues, we can protect the result.

Finally, the Atom community would benefit from the resources and experience of people who have made Internet and Web specifications and standards before, in areas from QA to specification language to modularity and so forth.

As we do this, I think it's critical for people to realise that going to a standards body to produce a specification is NOT a guarantee of success; indeed, there are innumerable examples of efforts where people tried to "boil the ocean." The Atom Community will need to be incredibly pragmatic and focus on commoditising what it can agree upon to have a chance of being successful, wherever it goes.

In this fashion, you're absolutely right - we shouldn't be looking for momentum, in the sense that there will continue to be rapid change in the spec itself. Rather, we'll be coming to agreement on the parts that we've already talked about, if we're doing the right thing. Personally, I think time time is ripe for this to happen to the Atom core.


On May 13, 2004, at 4:42 PM, Micone, Andrew (Manpower Contract) wrote:


Stepping back for a minute, would it be a better option to let Atom
mature more and gain wider adoption before submitting it to a standards
process? I can't speak for IETF's or W3C's process, but I've noticed in
my involvement with other XML standards bodies that while they are great
for stabilizing a standard, they tend to be the death of momentum. Just
a thought.

-----Original Message-----
From: owner-atom-syntax@xxxxxxxxxxxx
[mailto:owner-atom-syntax@xxxxxxxxxxxx] On Behalf Of Tim Bray
Sent: Thursday, May 13, 2004 4:35 PM
To: Atom Syntax; iesg@xxxxxxxx
Subject: Re: W3C response to proposed Atom Publishing Format and
Protocol (atompub) working group



On May 13, 2004, at 8:41 AM, Eric Miller wrote:

Due to its close relationship to the range of Web-based technologies,
we suggest that the Atom community propose to do this work in the W3C.

My take on this.

Advantages of the W3C:
- Web-centric, which Atom is
- good IPR policies
- one person getting paid to work on Atom understands the process (me)

Disadvantages of the W3C:
- many active Atom people don't work for W3C members

Advantages of the IETF:
- Open to all who participate
- Predisposition to do most of work by email
- "Rough consensus and running code"

Disadvantages of the IETF:
- No long-term Atom community members really understand the process

Personally, I think we could get a good result in either organization
and would not push back against the community if many voices were
raised in favor of switching horses to the W3C.  Looking at the pros
and cons above, my leaning is towards the IETF.  I tend not to be too
worried about intellectual-property issues on this one, since
syndication technology has been around for a long long time, back to
the days of 110-baud modems; it's really hard to imagine a credible
claim lurking out there.

-Tim



--
Mark Nottingham     http://www.mnot.net/